arXiv subjects

Rong Zhao

Publications and source records attributed to Rong Zhao.

At least 19 recordsLinked to original sources

arXiv2026

Mapping the City Through the Lens of Language Models

Language models often complete an underspecified reference to a city with unstated assumptions about urban size, form, infrastructure, environment, and function. We measure those assumptions without naming places. Ten open-weight checkpoints rate anonymized profiles derived from real morphological urban centres across 40 audited indicators and seven domains. The design combines constrained probability-based ratings, prespecified reliability screens, lineage-aware aggregation, multiple population weightings, an independent replication sample, and whole-profile validation. The clearest shared tendency favours urban profiles with larger developed area, faster recent growth, greater mapped infrastructure and non-residential capacity, and less sparse form. Most eligible directions recur in the replication data, and direct ratings of complete profiles show moderate agreement with the indicator-wise construction. Geographic differences shrink after accounting for city scale and development, while reliably measured paired tasks indicate that typicality and desirability are often closely aligned. The framework makes an otherwise vague notion of what models regard as an ordinary city empirically traceable. The resulting evidence delineates a shared yet model-dependent portrait of the city through the lens of language models.

cs.CL↗

arXiv2026

Mind Dreamer: Untethering Imagination via Active Causal Intervention on Latent Manifolds

Model-Based Reinforcement Learning yields sample efficiency via latent imagination, yet remains constrained by Historical Tethering: imagination is typically initialized from observed states. This creates a learning asymmetry, where the world model's manifold discovery outpaces the policy's sparse-reward optimization. We propose Mind Dreamer (MD), a framework that instantiates Active Causal Intervention to transcend Markovian continuity. MD reformulates discovery as the minimization of a global Relay Expected Free Energy. Instead of initializing from historical data, it draws initial states from an adversarial generator $s_0 \sim p_{gen}(\cdot)$, creating non-continuous latent jumps to epistemic blind spots that are physically plausible yet cognitively challenging. We derive Relay Value Function and Relay Uncertainty Function to resolve the credit assignment paradox across these spatial ruptures. Treating synthesized anchors as interventional intermediary states, these potentials propagate pragmatic and epistemic value through Bellman-style backups. Notably, we prove that uncertainty propagation across discontinuities necessitates a quadratic discount $\gamma^2$, establishing a formal epistemic horizon. Theoretically, MD approximates a variance-minimizing importance sampler that expands the manifold's spectral gap, reducing the hitting time to critical bottleneck states. Empirically, MD achieves a 1.67$\times$ average speedup over DreamerV3 on DeepMind Control Suite, reaching 8.8$\times$ in sparse-reward tasks.

cs.LG↗

arXiv2026

Joint sparse coding and temporal dynamics support context reconfiguration

Adaptive behavior requires the brain to transition between distinct contexts while maintaining representations of prior experience. The ability to reconfigure neural representations without erasing previously acquired knowledge is central to learning in dynamic environments, yet the neural mechanisms that support this balance remain unclear. Understanding these mechanisms is also critical for addressing catastrophic forgetting in artificial systems designed for lifelong learning. Here, we identify joint sparse coding and temporal dynamics in both the mouse medial prefrontal cortex (mPFC) and computational networks as mechanisms that help preserve prior representations during context transitions. Specifically, sparsity in context-dependent representations reduces cross-context interference, whereas temporal dynamics within the network activity further enhance context separability across time. Strikingly, networks endowed with both properties, such as spiking neural networks, exhibit improved retention during lifelong learning without auxiliary heuristics. These findings establish joint sparse coding and temporal dynamics as a core mechanism supporting flexible context reconfiguration in lifelong learning and, through their activity constraining nature, as an energy-efficient architectural principle for stable adaptation. Together, they provide a mechanistic framework for understanding how the brain preserves prior knowledge while flexibly adapting to new contexts.

q-bio.NC↗

arXiv2026

Culturally uneven urban perception in large language models

Large language models (LLMs) are increasingly used to describe and evaluate cities, yet the cultural structure of their urban judgments remains understudied. Here we introduce a measurement framework for testing whether LLM-based urban perception is culturally neutral, using a globally stratified street-view image dataset. Open-ended descriptions and structured scores generated by three frontier multimodal models all show that the neutral baseline lies closer to regional framings associated with Europe and North America than to other cultural framings. Comparisons between AI and human urban perception further show that prompting can move AI responses closer to specific regional human descriptions, but fails to recover the variety and diversity of human responses, flattening observed demographic patterns and introducing sentiment-based self-favouring bias. These results indicate a systematic risk in treating AI as a neutral tool for urban tasks, especially when model outputs are used to compare, evaluate or represent cities across cultural contexts.

cs.CL↗

arXiv2026

Spatio-Temporal Difference Guided Motion Deblurring with the Complementary Vision Sensor

Motion blur arises when rapid scene changes occur during the exposure period, collapsing rich intra-exposure motion into a single RGB frame. Without explicit structural or temporal cues, RGB-only deblurring is highly ill-posed and often fails under extreme motion. Inspired by the human visual system, brain-inspired vision sensors introduce temporally dense information to alleviate this problem. However, event cameras still suffer from event rate saturation under rapid motion, while the event modality entangles edge features and motion cues, which limits their effectiveness. As a recent breakthrough, the complementary vision sensor (CVS), Tianmouc, captures synchronized RGB frames together with high-frame-rate, multi-bit spatial difference (SD, encoding structural edges) and temporal difference (TD, encoding motion cues) data within a single RGB exposure, offering a promising solution for RGB deblurring under extreme dynamic scenes. To fully leverage these complementary modalities, we propose Spatio-Temporal Difference Guided Deblur Net (STGDNet), which adopts a recurrent multi-branch architecture that iteratively encodes and fuses SD and TD sequences to restore structure and color details lost in blurry RGB inputs. Our method outperforms current RGB or event-based approaches in both synthetic CVS dataset and real-world evaluations. Moreover, STGDNet exhibits strong generalization capability across over 100 extreme real-world scenarios. Project page: https://tmcDeblur.github.io/

cs.CV↗

arXiv2026

Does Entry of Food-and-Drink Establishments Raise Local House Prices? Event-Study Evidence from London

Restaurants, cafes, pubs, and takeaways are among the most visible markers of neighborhood change, yet whether their arrival is capitalised into nearby housing values remains empirically unsettled. We assemble a London-wide panel linking Land Registry prices, non-domestic EPC lodgement timings for food-and-drink establishments, and neighborhood amenity measures at the LSOA level. Our preferred annual event-study design defines treatment as the first clean-onset year in which an LSOA records at least two eligible EPC lodgements for food-and-drink establishments, after a two-year lookback with no prior entries. In this specification, pre-trend tests are not rejected in either the stacked or Sun-Abraham estimators, and log house prices rise gradually from about 0.5% in the event year to roughly 3.4--3.7% by years four and five. The results are consistent with local amenity capitalization following commercial entry, while remaining appropriately cautious about endogenous siting and concurrent redevelopment.

physics.soc-ph↗

arXiv2026

UrbanAlign: Post-hoc Semantic Calibration for VLM-Human Preference Alignment

Vision-language models (VLMs) can describe urban scenes in rich detail, yet consistently fail to produce reliable human preference labels in domain-specific tasks such as safety assessment and aesthetic evaluation. The standard fix, fine-tuning or RLHF, requires large-scale annotations and model retraining. We ask a different question: can a frozen VLM be aligned with human preferences without modifying any weights? Our key insight is that VLMs are strong concept extractors but poor decision calibrators. We propose a three-stage post-hoc pipeline that exploits this asymmetry: (i) interpretable evaluation dimensions are automatically mined from consensus exemplars; (ii) an Observer-Debater-Judge chain extracts robust concept scores from the frozen VLM; and (iii) locally-weighted ridge regression on a hybrid manifold calibrates these scores to human ratings. Applied as UrbanAlign on Place Pulse 2.0, the framework reaches 72.2% accuracy (kappa=0.45) across six perception categories, outperforming all baselines by +11.0 pp and zero-shot VLM by +15.5 pp, with full interpretability and zero weight modification.

cs.CV↗

arXiv2025

TopoPerception: A Shortcut-Free Evaluation of Global Visual Perception in Large Vision-Language Models

Large Vision-Language Models (LVLMs) typically align visual features from an encoder with a pre-trained Large Language Model (LLM). However, this makes the visual perception module a bottleneck, which constrains the overall capabilities of LVLMs. Conventional evaluation benchmarks, while rich in visual semantics, often contain unavoidable local shortcuts that can lead to an overestimation of models' perceptual abilities. Here, we introduce TopoPerception, a benchmark that leverages topological properties to rigorously evaluate the global visual perception capabilities of LVLMs across various granularities. Since topology depends on the global structure of an image and is invariant to local features, TopoPerception enables a shortcut-free assessment of global perception, fundamentally distinguishing it from semantically rich tasks. We evaluate state-of-the-art models on TopoPerception and find that even at the coarsest perceptual granularity, all models perform no better than random chance, indicating a profound inability to perceive global visual features. Notably, a consistent trend emerge within model families: more powerful models with stronger reasoning capabilities exhibit lower accuracy. This suggests that merely scaling up models is insufficient to address this deficit and may even exacerbate it. Progress may require new training paradigms or architectures. TopoPerception not only exposes a critical bottleneck in current LVLMs but also offers a lens and direction for improving their global visual perception. The data and code are publicly available at: https://github.com/Wenhao-Zhou/TopoPerception.

cs.AI↗

arXiv2025

A New Targeted-Federated Learning Framework for Estimating Heterogeneity of Treatment Effects: A Robust Framework with Applications in Aging Cohorts

Analyzing data from multiple sources offers valuable opportunities to improve the estimation efficiency of causal estimands. However, this analysis also poses many challenges due to population heterogeneity and data privacy constraints. While several advanced methods for causal inference in federated settings have been developed in recent years, many focus on difference-based averaged causal effects and are not designed to study effect modification. In this study, we introduce a novel targeted-federated learning framework to study the heterogeneity of treatment effects (HTEs) for a targeted population by proposing a projection-based estimand. This HTE framework integrates information from multiple data sources without sharing raw data, while accounting for covariate distribution shifts among sources. Our proposed approach is shown to be doubly robust, conveniently supporting both difference-based estimands for continuous outcomes and odds ratio-based estimands for binary outcomes. Furthermore, we develop a communication-efficient bootstrap-based selection procedure to detect non-transportable data sources, thereby enhancing robust information aggregation without introducing bias. The superior performance of the proposed estimator over existing methods is demonstrated through extensive simulation studies, and the utility of our approach has been shown in a real-world data application using nationwide Medicare-linked data.

stat.ME↗

arXiv2025

CompAir: Synergizing Complementary PIMs and In-Transit NoC Computation for Efficient LLM Acceleration

The rapid advancement of Large Language Models (LLMs) has revolutionized various aspects of human life, yet their immense computational and energy demands pose significant challenges for efficient inference. The memory wall, the growing processor-memory speed disparity, remains a critical bottleneck for LLM. Process-In-Memory (PIM) architectures overcome limitations by co-locating compute units with memory, leveraging 5-20$\times$ higher internal bandwidth and enabling greater energy efficiency than GPUs. However, existing PIMs struggle to balance flexibility, performance, and cost-efficiency for LLMs' dynamic memory-compute patterns and operator diversity. DRAM-PIM suffers from inter-bank communication overhead despite its vector parallelism. SRAM-PIM offers sub-10ns latency for matrix operation but is constrained by limited capacity. This work introduces CompAir, a novel PIM architecture that integrates DRAM-PIM and SRAM-PIM with hybrid bonding, enabling efficient linear computations while unlocking multi-granularity data pathways. We further develop CompAir-NoC, an advanced network-on-chip with an embedded arithmetic logic unit that performs non-linear operations during data movement, simultaneously reducing communication overhead and area cost. Finally, we develop a hierarchical Instruction Set Architecture that ensures both flexibility and programmability of the hybrid PIM. Experimental results demonstrate that CompAir achieves 1.83-7.98$\times$ prefill and 1.95-6.28$\times$ decode improvement over the current state-of-the-art fully PIM architecture. Compared to the hybrid A100 and HBM-PIM system, CompAir achieves 3.52$\times$ energy consumption reduction with comparable throughput. This work represents the first systematic exploration of hybrid DRAM-PIM and SRAM-PIM architectures with in-network computation capabilities, offering a high-efficiency solution for LLM.

cs.AR↗

arXiv2025

JUNO 20-inch PMT and electronics system characterization using large pulses of PMT dark counts at the Pan-Asia testing platform

The main goal of the JUNO experiment is to determine the neutrino mass ordering with a 20kt liquid-scintillator detector. The 20-inch PMT and its 1F3 (one for three) electronics are crucial to realize the excellent energy resolution of at least 3% at 1MeV. The knowledge on the PMT and 1F3 electronics response is critical for detector performance understanding. A study of the JUNO 20-inch PMT and 1F3 electronics system characterization is presented using large pulses of PMT dark count at the Pan-Asia testing platform in China. Thanks to its broad amplitude range and high rate, the large pulse signals are also used to investigate the PMT after pulse response.

physics.ins-det↗

arXiv2025

Dark Count Rate Stability of JUNO 20-inch PMTs in Mass Testing

The Jiangmen Underground Neutrino Observatory (JUNO) is an ambitious multipurpose neutrino experiment designed to determine the neutrino mass ordering, with an impressive energy resolution goal of at least 3% at 1 MeV. To achieve a photon detection coverage of approximately 75%, JUNO will utilize two types of 20-inch photomultiplier tubes (PMTs): the large PMT (LPMT) and the microchannel plate PMT (MCP-PMT). A significant concern in high-precision neutrino measurements is the dark count rate (DCR) of PMTs, which introduces noise that can adversely affect energy measurement accuracy. During the mass testing phase of the JUNO 20-inch PMTs, comprehensive measurements of the DCR were undertaken. These measurements not only captured the DCR values of individual PMTs but also examined the stability and temperature dependence of the DCR at an operating gain of (1x10^7). This paper presents a detailed characterization of the DCR of the JUNO 20-inch PMTs, investigating factors such as cooling time, temperature variations, and long-term stability using the JUNO Pan-Asia PMT testing facilities. The results reveal distinct DCR characteristics between the two types of PMTs, providing valuable insights into the nature of DCR and its implications for JUNO's scientific objectives. In addition to performance characterization, we implemented a monitoring system to track DCR stability over time. Notably, several spikes in DCR were identified, prompting a preliminary investigation into their causes. Potential factors contributing to these spikes, such as flasher events, were explored using coincidence rate analysis and complementary imaging techniques. The findings from this study are crucial for optimizing the performance of PMTs in JUNO, ultimately aiding the experiment in achieving its goals related to neutrino physics.

physics.ins-det↗

arXiv2025

Topology-Aware Virtualization over Inter-Core Connected Neural Processing Units

With the rapid development of artificial intelligence (AI) applications, an emerging class of AI accelerators, termed Inter-core Connected Neural Processing Units (NPU), has been adopted in both cloud and edge computing environments, like Graphcore IPU, Tenstorrent, etc. Despite their innovative design, these NPUs often demand substantial hardware resources, leading to suboptimal resource utilization due to the imbalance of hardware requirements across various tasks. To address this issue, prior research has explored virtualization techniques for monolithic NPUs, but has neglected inter-core connected NPUs with the hardware topology. This paper introduces vNPU, the first comprehensive virtualization design for inter-core connected NPUs, integrating three novel techniques: (1) NPU route virtualization, which redirects instruction and data flow from virtual NPU cores to physical ones, creating a virtual topology; (2) NPU memory virtualization, designed to minimize translation stalls for SRAM-centric and NoC-equipped NPU cores, thereby maximizing the memory bandwidth; and (3) Best-effort topology mapping, which determines the optimal mapping from all candidate virtual topologies, balancing resource utilization with end-to-end performance. We have developed a prototype of vNPU on both an FPGA platform (Chipyard+FireSim) and a simulator (DCRA). Evaluation results indicate that, compared to other virtualization approaches such as unified virtual memory and MIG, vNPU achieves up to a 2x performance improvement across various ML models, with only 2% hardware cost.

cs.AR↗

arXiv2025

GenAI Models Capture Urban Science but Oversimplify Complexity

Generative artificial intelligence (GenAI) models are increasingly used for scientific data generation, yet their alignment with empirical knowledge in urban science remains unclear. We therefore ask whether generated urban data reproduce empirical regularities and support repeatable experiments. We introduce AI4US, a framework for evaluating data synthesis and conditional intervention across text and image modalities. Four cases spanning urban systems, within-city structure, neighbourhood vitality and streetscape perception are evaluated against published parameters, observed urban data and human judgements. Generated outputs recovered recognizable scaling and distance-decay patterns and yielded measurable associations between neighbourhood morphology indicators and pedestrian activity, while model judgements showed positive agreement with sampled human choices. Controlled changes to urban conditions produced repeatable output responses across all four cases. However, generated data often compressed empirical numerical coverage, local variation and visual diversity. In an inspectable GenAI model, intermediate activations linearly distinguished some urban relationships, while activation edits changed only selected output scores. AI4US provides an empirically grounded approach for assessing GenAI as the virtual urban laboratory in urban data synthesis and controlled model experiments, while revealing its tendency to simplify empirical complexity.

physics.soc-ph↗

arXiv2025

Physics-Driven Local-Whole Elastic Deformation Modeling for Point Cloud Representation Learning

Existing point cloud representation learning methods primarily rely on data-driven strategies to extract geometric information from large amounts of scattered data. However, most methods focus solely on the spatial distribution features of point clouds while overlooking the relationship between local information and the whole structure, which limits the accuracy of point cloud representation. Local information reflect the fine-grained variations of an object, while the whole structure is determined by the interaction and combination of these local features, collectively defining the object's shape. In real-world, objects undergo deformation under external forces, and this deformation gradually affects the whole structure through the propagation of forces from local regions, thereby altering the object's geometric features. Therefore, appropriately introducing a physics-driven mechanism to capture the topological relationships between local parts and the whole object can effectively mitigate for the limitations of data-driven point cloud methods in structural modeling, and enhance the generalization and interpretability of point cloud representations for downstream tasks such as understanding and recognition. Inspired by this, we incorporate a physics-driven mechanism into the data-driven method to learn fine-grained features in point clouds and model the structural relationship between local regions and the whole shape. Specifically, we design a dual-task encoder-decoder framework that combines the geometric modeling capability of data-driven implicit fields with physics-driven elastic deformation. Through the integration of physics-based loss functions, the framework is guided to predict localized deformation and explicitly capture the correspondence between local structural changes and whole shape variations.

cs.CV↗

arXiv2025

Quantitative evaluation of brain-inspired vision sensors in high-speed robotic perception

Perception systems in robotics encounter significant challenges in high-speed and dynamic conditions when relying on traditional cameras, where motion blur can compromise spatial feature integrity and task performance. Brain-inspired vision sensors (BVS) have recently gained attention as an alternative, offering high temporal resolution with reduced bandwidth and power requirements. Here, we present the first quantitative evaluation framework for two representative classes of BVSs in variable-speed robotic sensing, including event-based vision sensors (EVS) that detect asynchronous temporal contrasts, and the primitive-based sensor Tianmouc that employs a complementary mechanism to encode both spatiotemporal changes and intensity. A unified testing protocol is established, including crosssensor calibrations, standardized testing platforms, and quality metrics to address differences in data modality. From an imaging standpoint, we evaluate the effects of sensor non-idealities, such as motion-induced distortion, on the capture of structural information. For functional benchmarking, we examine task performance in corner detection and motion estimation under different rotational speeds. Results indicate that EVS performs well in highspeed, sparse scenarios and in modestly fast, complex scenes, but exhibits performance limitations in high-speed, cluttered settings due to pixel-level bandwidth variations and event rate saturation. In comparison, Tianmouc demonstrates consistent performance across sparse and complex scenarios at various speeds, supported by its global, precise, high-speed spatiotemporal gradient samplings. These findings offer valuable insights into the applicationdependent suitability of BVS technologies and support further advancement in this area.

cs.RO↗

arXiv2025

Mapping at First Sense: A Lightweight Neural Network-Based Indoor Structures Prediction Method for Robot Autonomous Exploration

Autonomous exploration in unknown environments is a critical challenge in robotics, particularly for applications such as indoor navigation, search and rescue, and service robotics. Traditional exploration strategies, such as frontier-based methods, often struggle to efficiently utilize prior knowledge of structural regularities in indoor spaces. To address this limitation, we propose Mapping at First Sense, a lightweight neural network-based approach that predicts unobserved areas in local maps, thereby enhancing exploration efficiency. The core of our method, SenseMapNet, integrates convolutional and transformerbased architectures to infer occluded regions while maintaining computational efficiency for real-time deployment on resourceconstrained robots. Additionally, we introduce SenseMapDataset, a curated dataset constructed from KTH and HouseExpo environments, which facilitates training and evaluation of neural models for indoor exploration. Experimental results demonstrate that SenseMapNet achieves an SSIM (structural similarity) of 0.78, LPIPS (perceptual quality) of 0.68, and an FID (feature distribution alignment) of 239.79, outperforming conventional methods in map reconstruction quality. Compared to traditional frontier-based exploration, our method reduces exploration time by 46.5% (from 2335.56s to 1248.68s) while maintaining a high coverage rate (88%) and achieving a reconstruction accuracy of 88%. The proposed method represents a promising step toward efficient, learning-driven robotic exploration in structured environments.

cs.RO↗

arXiv2025

New Stress-dependent Elastic Wave Velocity Models for Reservoir Rocks with Applications

This study presents new elastic velocity-effective stress laws for reservoir rocks. These models are grounded in previously established correlations between elastic modulus and porosity, which incorporate critical porosity. The accuracy of the models is validated against wave velocities from 38 core samples, yielding coefficients of determination ($\mathrm{R}^2$) of 0.9994 for compressional wave and 0.9985 for shear wave. A sensitivity analysis reveals that the maximum uncertainties for compressional and shear waves are less than $\pm$5.5% and $\pm$7.5%, respectively. To demonstrate the applicability of the proposed models, a case study was conducted on three wells in the Northern Carnarvon Basin, where the new elastic wave velocity-effective stress laws produced reliable predictions for velocity logs in the studied formations. The relationships reported herein may prove beneficial for hydrocarbon exploration, production, and ensuring drilling safety in both unconventional and conventional fields.

physics.geo-ph↗